 Welcome back and this video we will talk about what are the tools or used in ML infrastructure in the companies. So if you are you know interested in more about tools or in the ML domains or you want to get a you know to join a job as a entry level data scientist in our industry. So we will look at what is needed. So I would suggest start reading more books, the very good machine learning books available if you are interested in programming and books, good books available for Python versus machine learning something like that. You are worried about what is a good book, what are the best book I have to read, don't worry go and check in our internet, there are plenty of advice for you and look at the no suggested books, pick the second or third book they suggested, it's always works well. Okay, some of the well-known books like Professor Andrea and she wrote a book from based on his course, machine learning earning or Christopher Bishop wrote a very book on pattern, pattern recognition in ML. So it is based on what is your interest, if you are interested in learning machine learning in programming languages, script language like Python or R then choose the books. If you are interested more about the mathematical aspects or it is happening, choose you know books which are serve that purpose. Also learn a lot about better tools, don't stop with you know orange or Vika, they are not enough if you want looking for jobs so more tools will be good. So based on talk to my friend who is you know data analyst in you know multinational company in abroad, I asked him what are the tools you have to use, what are the tools they have to learn for entry level analyst in machine learning, what will help get the students to get a job in industry. So I said that learn Python or R it is very, very important you know it is a scripting language everybody is moving towards Python or R and if you have you know good skill on C++ C sharp or in a Java still preferred but Python R is what you taking up you know. Use ID for coding you know some you know interactive development environment any environment which supports Python R is what is adjusted to it because that helps a lot to learn coding. If you don't have anything you don't have resource just use the Google online resource Google Colab. That is it if you have heavy computation right so you want to apply you know machine learning like a neural networks on millions of data or your multiple layers you don't have system with you you can upload the data in Google Colab and do it that is very good. The only thing is make sure that you are upgrading the data the data have a you know ID identified anonymized you don't give the data to Google you know that is important. So the tools is adjusted go beyond you know the orange or record rapid reminder go for power BI, tab view kind of tools and that tools helps you to you know understand work with the large big data kind of thing. The data with the tools we use okay good for you know small low size of data if in industry they use data from you know millions of customers and thousands of customers and they turn out this data to predict something good. So use the tools like power BI and tab view will be helpful. How to learn that you know most of these tools are available commercial for academic purpose for free but you might get a free license version for some you know some restricted period of time. There are a lot of videos available how to use tab view or power BI in the YouTube. For features extraction is a data robot H2 you know H2O driver let us say et cetera but the point is it is all commercial. For you know for a non-commercial purpose we talked about that one feature tools that is feature extractor you know week 2 we gave you know one particular tool. So there is a non-commercial open source tools also available. If you do not find you know there is a commercial I do not know what is alternative go to a website called alternative 2 or type alternative 2 data robot in Google that website will come up in the past. It will suggest all the other alternatives with the visit of open source visit a commercial they give the data to you. So pick those you know tool and learn more about it. So if you do not know where to find you know open source alternative of this particular type in Google alternative to data robot. Look for website it is called alternative dot something you know some domain and they will describe you the actually which is other alternatives which is open source or commercial. More important thing in industry is not just you know collected or extract features and apply some ML on tool beyond that is interpretation. So how do you interpret the model? So they all go towards you know decision trees or linear regressions to interpret the model you know explainable AI. So use the tools called interpret ML it is very very important to interpret the models and they help you to do that. So these are the tools suggestions from you know my friend at industry he said that it is useful if you are looking for job in you know in data science career. So and he suggested virtualization plotly in a python it has a dash app, dash app is generated by a developer university which creates interactive dashboards. This plotly is actually have this dash app also included so it is all in private then. So if you want to create interactive things it is all easily available there. So if you do not know there is no this tools are tough how do I this is one person's view is it a true or not. If you want to know what is the current thing in the AI or ML industry, I would say this go to type this words like type things called AI or ML infrastructure that is it we type it look at the first one. Let us see that detail see there is a data preparation model building production you might want to look at the website and they will explain each and everything and they will explain why it is used. I was telling about you know it is too much big data you can use this it is all data preparation and exploration process and feature labs or feature tools. The feature tools is kind of open source one just use it and the model building you can use multiple ways like tensor flow or you know different model builders this is all things available. So go and look at the current things and this is always good you know you can go and look for latest things from the Google. So what I am saying is if you are really interested to take a career in a data science just look for this tools which is open source and learn them and not all of them there is nobody is going to be nobody is going to learn everything in this. Pick one tool in each section and learn about them if you have time pick one or two more tools in each step that is data preparation and modeling and interpretation and integrating the tool with the industry is important it is not that interpreting how this is coming back to the industry how they are giving feedback information to the new knowledge. Learn about this one or tools in each step pick it up and create a resume and apply data from Kaggle or any online resources pick it up. This will you are then applied in different domains right the Kaggle gives data from different domains. This will give you a good set of you know your resume to get hired from the good companies like data science companies that is the idea ok. So yeah so I suggest some tools by some of my friends suggestion or you can go and check in the online for the latest tools. So thank you for taking this course I hope you really enjoyed and you learn something and thanks for being with us for last 12 weeks and thanks for doing all assignments I hope you learn some tools and you take something from this course like you know I want to do some research in LA or I want to learn something in ML whatever this direction you take we are very happy for that. And if you are anything you know for the questions or something talk to your peers or learners in the discussion forum ask questions ask learners to respond or you know discuss with the learners or do leachers if you have any questions you know specific questions how to do that. As I mentioned most of the questions can be answered by Google already there is nothing else that we know better than you know what the world knows. So but keep learning and I hope you really enjoyed the course. Thank you.